Building taxonomy of web search intents for name entity queries

Abstract

A significant portion of web search queries are name entity queries. The major search engines have been exploring various ways to provide better user experiences for name entity queries, such as showing "search tasks" (Bing search) and showing direct answers (Yahoo!, Kosmix). In order to provide the search tasks or direct answers that can satisfy most popular user intents, we need to capture these intents, together with relationships between them.In this paper we propose an approach for building a hierarchical taxonomy of the generic search intents for a class of name entities (e.g., musicians or cities). The proposed approach can find phrases representing generic intents from user queries, and organize these phrases into a tree, so that phrases indicating equivalent or similar meanings are on the same node, and the parent-child relationships of tree nodes represent the relationships between search intents and their sub-intents. Three different methods are proposed for tree building, which are based on directed maximum spanning tree, hierarchical agglomerative clustering, and pachinko allocation model. Our approaches are purely based on search logs, and do not utilize any existing taxonomies such as Wikipedia. With the evaluation by human judges (via Mechanical Turk), it is shown that our approaches can build trees of phrases that capture the relationships between important search intents.

Mining latent entity structures from massive unstructured and interconnected data

“…Accuracy on the predicted parent y i (A par ), and the accuracy on the predicted relation pairs x ij (A pair ), are two most natural evaluation criteria; they or their variants (Precision, Recall, etc.) are employed by most previous studies [73,99,83]. However, such measures only evaluate the prediction variables on each node or each edge in an isolated view, missing some aspects of the comprehensive goodness of structure.…”

“…Kemp and Tenenbaum [45] propose to use a generative model to find structure in data, and Maiya and Berger-Wolf [57] apply the similar idea in inferring the social network hierarchy with the maximum likelihood of observing the interactions among the people. In the Web search domain, Yin and Shah [99] study the problem of building taxonomies of search intents for entity queries based on inferring the "belonging" relationships between them with unsupervised approaches. There is other related work for building taxonomies from Web tags with similar methodology [26].…”

Towards Semantic User Query: A Review

“…A substitute of measuring user intentions is to study user search queries and browsing behavior for text-based search (Broder, 2002;Jansen et al, 2008;Yin and Shah, 2010;Kumar and Tomkins, 2010) and image search (Kofler and Lux, 2009;Lux et al, 2010).…”

Learning search tasks in queries and web pages via graph regularization

“…Queries and web pages with different topics (such as computers or cars) naturally have different search tasks, thus making our task definition at a finer scale than the existing binary or three-class web search taxonomy. Yin et al [21] build a hierarchical taxonomy of the generic search intents for a class of named entities by analyzing the relationships between queries and grouping them into a tree structure, which is essentially data-driven. On the contrary, we focus on classifying queries and web pages into several pre-defined tasks that are of special interest among search engine users.…”

“…As discussed in [21], the content in a query can be divided into two parts: named entities 3 and other terms. In this paper, we define the search task to be the action 3 Here we mainly work with named entity queries and related web that the user wants to perform towards the entities.…”